[Bug]api-key is set up incorrectly #101
Description
here my command
genai-bench benchmark \
--api-backend sglang \
--task text-to-text \
--api-model-name "DeepSeek-V3.1" \
--api-base https://172.10.2.3/v3.1 \
--api-key "112233445566" \
--additional-request-params '{"stream_options": {"include_usage": true}, "chat_template_kwargs": {"thinking": true}}' \
--server-engine "SGLang" \
--server-version "0.4.10" \
--server-gpu-type "H200" \
--server-gpu-count "32*8" \
--model-tokenizer /root/deepseek-tokenizer-v3.1 \
--experiment-base-dir ./genai-bench-result \
--dataset-path /root/LongAlpaca-12k.csv \
--dataset-prompt-column "instruction" \
--traffic-scenario "U(1,12001)/(1,2001)" \
--num-concurrency 260 \
--max-requests-per-run 3000 \
--max-time-per-run 60 \
--num-workers 2 \
--spawn-rate 20
run faled, here the log:
[11/04/25 10:49:41] INFO 👋 Welcome to genai-bench 0.0.2! I am an intelligent benchmark tool for Large Language Model. cli.py:154
INFO Options you provided: cli.py:160
INFO num_concurrency: (1,) cli.py:162
INFO batch_size: [1] cli.py:162
INFO api_backend: sglang cli.py:162
INFO task: text-to-text cli.py:162
INFO api_model_name: DeepSeek-V3.1 cli.py:162
INFO api_base: https://172.10.2.3/v3.1 cli.py:162
INFO api_key: 112233445566 cli.py:162
INFO additional_request_params: {'stream_options': {'include_usage': True}, 'chat_template_kwargs': {'thinking': True}} cli.py:162
INFO server_engine: SGLang cli.py:162
INFO server_version: 0.4.10 cli.py:162
INFO server_gpu_type: H200 cli.py:162
INFO server_gpu_count: 32*8 cli.py:162
INFO model_tokenizer: /root/deepseek-tokenizer-v3.1 cli.py:162
INFO experiment_base_dir: ./genai-bench-result cli.py:162
INFO dataset_path: /root/LongAlpaca-12k.csv cli.py:162
INFO dataset_prompt_column: instruction cli.py:162
INFO traffic_scenario: ['U(1,12001)/(1,2001)'] cli.py:162
INFO max_requests_per_run: 1 cli.py:162
INFO max_time_per_run: 60 cli.py:162
INFO num_workers: 1 cli.py:162
INFO spawn_rate: 20 cli.py:162
INFO gcp_credentials_path: None cli.py:162
INFO gcp_location: us-central1 cli.py:162
INFO gcp_project_id: None cli.py:162
INFO azure_ad_token: None cli.py:162
INFO azure_api_version: 2024-02-01 cli.py:162
INFO azure_deployment: None cli.py:162
INFO azure_endpoint: None cli.py:162
INFO aws_region: None cli.py:162
INFO aws_profile: None cli.py:162
INFO aws_session_token: None cli.py:162
INFO aws_secret_access_key: None cli.py:162
INFO aws_access_key_id: None cli.py:162
INFO model_api_key: None cli.py:162
INFO model_auth_type: None cli.py:162
INFO region: None cli.py:162
INFO security_token: None cli.py:162
INFO profile: DEFAULT cli.py:162
INFO config_file: ~/.oci/config cli.py:162
INFO auth: user_principal cli.py:162
INFO model: deepseek-tokenizer-v3.1 cli.py:162
INFO iteration_type: num_concurrency cli.py:162
INFO experiment_folder_name: None cli.py:162
INFO dataset_image_column: None cli.py:162
INFO dataset_config: None cli.py:162
INFO master_port: 5557 cli.py:162
INFO upload_results: False cli.py:162
INFO namespace: None cli.py:162
INFO github_repo: None cli.py:162
INFO github_owner: None cli.py:162
INFO github_token: None cli.py:162
INFO storage_gcp_credentials_path: None cli.py:162
INFO storage_gcp_project_id: None cli.py:162
INFO storage_azure_sas_token: None cli.py:162
INFO storage_azure_connection_string: None cli.py:162
INFO storage_azure_account_key: None cli.py:162
INFO storage_azure_account_name: None cli.py:162
INFO storage_aws_profile: None cli.py:162
INFO storage_aws_region: None cli.py:162
INFO storage_aws_session_token: None cli.py:162
INFO storage_aws_secret_access_key: None cli.py:162
INFO storage_aws_access_key_id: None cli.py:162
INFO storage_auth_type: None cli.py:162
INFO storage_prefix: cli.py:162
INFO storage_bucket: None cli.py:162
INFO storage_provider: oci cli.py:162
INFO Using sglang authentication cli.py:237
[11/04/25 10:49:42] INFO The average character token ratio is: 4.10597519729425 cli.py:262
INFO Loading CSV file: /root/LongAlpaca-12k.csv sources.py:69
[11/04/25 10:49:45] INFO Loaded 12000 rows from CSV file with columns: ['instruction', 'output', 'file', 'input'] sources.py:73
INFO This experiment will be saved in folder /root/genai-bench-result/sglang_SGLang_0.4.10_text-to-text_deepseek-tokenizer-v3.1_20251104_104945. cli.py:310
INFO bench-vm_4533774c69e045939838207046ac1d05 (index 0) reported as ready. 1 workers connected. runners.py:1019
[11/04/25 10:49:47] INFO Starting benchmark with concurrency=1, spawn_rate=20 cli.py:412
INFO Sending spawn jobs of 1 users at 20.00 spawn rate to 1 ready workers runners.py:746
INFO [Worker 0] Worker 0 started successfully and connected to master runner.py:182
INFO All users spawned: {"OpenAIUser": 1} (1 total users) runners.py:833
[11/04/25 10:49:48] INFO [Worker 0] ======================url=https://172.10.2.3/v3.1/v1/chat/completions,headers={'Authorization': "Bearer {'api_key': runner.py:182
'112233445566'}", 'Content-Type': 'application/json'}=====================
WARNING [Worker 0] Received error response from server. Error code: 401, message: {"error":{"message":"无效的令牌 (request id: 2025110410494817548450709760043)","type":"my_gw_error"}}. runner.py:182
INFO [Worker 0] ======================url=https://172.10.2.3/v3.1/v1/chat/completions,headers={'Authorization': "Bearer {'api_key': runner.py:182
'112233445566'}", 'Content-Type': 'application/json'}=====================
WARNING [Worker 0] Received error response from server. Error code: 401, message: {"error":{"message":"无效的令牌 (request id: 2025110410494829539122892188686)","type":"my_gw_error"}}. runner.py:182
INFO [Worker 0] ======================url=https://172.10.2.3/v3.1/v1/chat/completions,headers={'Authorization': "Bearer {'api_key': runner.py:182
'112233445566'}", 'Content-Type': 'application/json'}=====================
WARNING [Worker 0] Received error response from server. Error code: 401, message: {"error":{"message":"无效的令牌 (request id: 2025110410494842845579422772361)","type":"my_gw_error"}}. runner.py:182
INFO [Worker 0] ======================url=https://172.10.2.3/v3.1/v1/chat/completions,headers={'Authorization': "Bearer {'api_key': runner.py:182
'112233445566'}", 'Content-Type': 'application/json'}=====================
WARNING [Worker 0] Received error response from server. Error code: 401, message: {"error":{"message":"无效的令牌 (request id: 2025110410494855240007358256515)","type":"my_gw_error"}}. runner.py:182
INFO [Worker 0] ======================url=https://172.10.2.3/v3.1/v1/chat/completions,headers={'Authorization': "Bearer {'api_key': runner.py:182
'112233445566'}", 'Content-Type': 'application/json'}=====================
WARNING [Worker 0] Received error response from server. Error code: 401, message: {"error":{"message":"无效的令牌 (request id: 2025110410494867523575411824994)","type":"my_gw_error"}}. runner.py:182
INFO [Worker 0] ======================url=https://172.10.2.3/v3.1/v1/chat/completions,headers={'Authorization': "Bearer {'api_key': runner.py:182
'112233445566'}", 'Content-Type': 'application/json'}=====================
WARNING [Worker 0] Received error response from server. Error code: 401, message: {"error":{"message":"无效的令牌 (request id: 2025110410494876735453645882633)","type":"my_gw_error"}}. runner.py:182
INFO [Worker 0] ======================url=https://172.10.2.3/v3.1/v1/chat/completions,headers={'Authorization': "Bearer {'api_key': runner.py:182
'112233445566'}", 'Content-Type': 'application/json'}=====================
[11/04/25 10:49:49] INFO ⏩ Exit the run as 7 requests have been completed. utils.py:46
WARNING [Worker 0] Received error response from server. Error code: 401, message: {"error":{"message":"无效的令牌 (request id: 2025110410494892529093883715967)","type":"my_gw_error"}}. runner.py:182
INFO [Worker 0] ======================url=https://172.10.2.3/v3.1/v1/chat/completions,headers={'Authorization': "Bearer {'api_key': runner.py:182
'112233445566'}", 'Content-Type': 'application/json'}=====================
WARNING [Worker 0] Received error response from server. Error code: 401, message: {"error":{"message":"无效的令牌 (request id: 2025110410494899923319380065588)","type":"my_gw_error"}}. runner.py:182
INFO [Worker 0] ======================url=https://172.10.2.3/v3.1/v1/chat/completions,headers={'Authorization': "Bearer {'api_key': runner.py:182
'112233445566'}", 'Content-Type': 'application/json'}=====================
INFO bench-vm_4533774c69e045939838207046ac1d05 (index 0) reported that it has stopped runners.py:1082
INFO bench-vm_4533774c69e045939838207046ac1d05 (index 0) reported as ready. 1 workers connected. runners.py:1019
[11/04/25 10:49:50] ERROR Uncaught exception
the api-key is set up incorrectly； should be set "Bearer 112233445566", but "Bearer {'api_key':'112233445566'}"